Correction of Downward Atmospheric Radiation and Land Surface Emissivity in Ground Temperature Measurements of Complex Terrain Surfaces

Abstract

Land surface temperature (LST) refers to the temperature of objects on the Earth’s surface. It plays a key role in surface energy exchange and is closely related to climate change, vegetation cover, hydrological processes, and ecological dynamics. As the most fundamental and traditional method for LST measurements, ground observation serves as the theoretical reference for evaluating alternative acquisition techniques. However, in complex mountain environments, rapid weather changes and local climate effects introduce uncertainty into ground-based LST measurements. Moreover, the synergistic impact of near-surface atmospheric radiation and land surface emissivity (LSE) on ground-based LST remains an underexplored factor contributing to this uncertainty. Therefore, this study employs the thermal radiative transfer equation to correct the LST data from thermal infrared (TIR) radiometers. In addition, the error induced by downwelling atmospheric longwave radiation (DALR) and LSE was quantified across various land cover types and seasons. The results for the study region, located in the southwest karst valley area, reveal an overall annual correction bias (corrected LST–uncorrected LST) of 0.3 K. Analysis shows this bias is strongly modulated by land cover and weather conditions. Notably, concrete surfaces exhibit the highest bias (mean of 0.58 K, peaking at 2.03 K), while the lowest is for grassland (0.16 K on average). Clear-sky conditions also led to a higher bias (0.28 K) than cloudy conditions (0.22 K). Seasonally, biases were greater in summer (0.42 K) than in winter (0.21 K). The maximum biases caused by DALR and LSE in this region, associated with the lowest LSE and the greatest difference between surface-emitted radiation and DALR events throughout the year, can reach 2.40 K. These quantified errors confirm that DALR and LSE are nonnegligible sources of uncertainty in ground-based LST measurement.